End of training

README.md

@@ -2,6 +2,8 @@
 license: cc-by-nc-4.0
 library_name: peft
 tags:
+- text-to-audio
+- sweet-dreambooths/black-eyed-peas-v1-piano-prompts
 - generated_from_trainer
 base_model: facebook/musicgen-melody-large
 model-index:
@@ -16,7 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/sanchit-gandhi/black-eyed-peas/runs/17sgxs8t)
 # black-eyed-peas-v1-crafted-prompt-3-epochs-text-only-piano-prompts
 
-This model is a fine-tuned version of [facebook/musicgen-melody-large](https://huggingface.co/facebook/musicgen-melody-large) on an unknown dataset.
+This model is a fine-tuned version of [facebook/musicgen-melody-large](https://huggingface.co/facebook/musicgen-melody-large) on the SWEET-DREAMBOOTHS/BLACK-EYED-PEAS-V1-PIANO-PROMPTS - DEFAULT dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.0910
+- Text Cosine Sim: 0.4673
+- Audio Cosine Sim: 0.5568
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 2.6879999999999997,
-    "eval_audio_cosine_sim": 0.5482687950134277,
-    "eval_loss": 3.092636823654175,
-    "eval_runtime": 967.8097,
+    "epoch": 2.9690721649484537,
+    "eval_audio_cosine_sim": 0.5567758679389954,
+    "eval_loss": 3.091003179550171,
+    "eval_runtime": 1153.212,
     "eval_samples": 12,
-    "eval_samples_per_second": 0.012,
-    "eval_steps_per_second": 0.012,
-    "eval_text_cosine_sim": 0.4583999812602997,
-    "total_flos": 247581948423984.0,
-    "train_loss": 9.755998475211006,
-    "train_runtime": 4065.4818,
-    "train_samples": 125,
-    "train_samples_per_second": 0.092,
-    "train_steps_per_second": 0.005
+    "eval_samples_per_second": 0.01,
+    "eval_steps_per_second": 0.01,
+    "eval_text_cosine_sim": 0.46733546257019043,
+    "total_flos": 212383098180048.0,
+    "train_loss": 9.849881066216362,
+    "train_runtime": 1793.8931,
+    "train_samples": 97,
+    "train_samples_per_second": 0.162,
+    "train_steps_per_second": 0.01
 }

eval_results.json

@@ -1,10 +1,10 @@
 {
-    "epoch": 2.6879999999999997,
-    "eval_audio_cosine_sim": 0.5482687950134277,
-    "eval_loss": 3.092636823654175,
-    "eval_runtime": 967.8097,
+    "epoch": 2.9690721649484537,
+    "eval_audio_cosine_sim": 0.5567758679389954,
+    "eval_loss": 3.091003179550171,
+    "eval_runtime": 1153.212,
     "eval_samples": 12,
-    "eval_samples_per_second": 0.012,
-    "eval_steps_per_second": 0.012,
-    "eval_text_cosine_sim": 0.4583999812602997
+    "eval_samples_per_second": 0.01,
+    "eval_steps_per_second": 0.01,
+    "eval_text_cosine_sim": 0.46733546257019043
 }

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 2.6879999999999997,
-    "total_flos": 247581948423984.0,
-    "train_loss": 9.755998475211006,
-    "train_runtime": 4065.4818,
-    "train_samples": 125,
-    "train_samples_per_second": 0.092,
-    "train_steps_per_second": 0.005
+    "epoch": 2.9690721649484537,
+    "total_flos": 212383098180048.0,
+    "train_loss": 9.849881066216362,
+    "train_runtime": 1793.8931,
+    "train_samples": 97,
+    "train_samples_per_second": 0.162,
+    "train_steps_per_second": 0.01
 }

trainer_state.json

@@ -1,172 +1,151 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.6879999999999997,
+  "epoch": 2.9690721649484537,
   "eval_steps": 30,
-  "global_step": 21,
+  "global_step": 18,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.128,
-      "grad_norm": 1.7691179513931274,
-      "learning_rate": 2.857142857142857e-05,
-      "loss": 10.1767,
+      "epoch": 0.16494845360824742,
+      "grad_norm": 1.61822509765625,
+      "learning_rate": 2.8333333333333332e-05,
+      "loss": 10.1055,
       "step": 1
     },
     {
-      "epoch": 0.256,
-      "grad_norm": 1.533618450164795,
-      "learning_rate": 2.7142857142857144e-05,
-      "loss": 10.1053,
+      "epoch": 0.32989690721649484,
+      "grad_norm": 1.7437396049499512,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 10.2368,
       "step": 2
     },
     {
-      "epoch": 0.384,
-      "grad_norm": 1.8578174114227295,
-      "learning_rate": 2.5714285714285714e-05,
-      "loss": 10.1344,
+      "epoch": 0.4948453608247423,
+      "grad_norm": 1.7318273782730103,
+      "learning_rate": 2.5e-05,
+      "loss": 10.0526,
       "step": 3
     },
     {
-      "epoch": 0.512,
-      "grad_norm": 1.893380880355835,
-      "learning_rate": 2.4285714285714288e-05,
-      "loss": 10.0509,
+      "epoch": 0.6597938144329897,
+      "grad_norm": 2.097566604614258,
+      "learning_rate": 2.3333333333333336e-05,
+      "loss": 10.1989,
       "step": 4
     },
     {
-      "epoch": 0.64,
-      "grad_norm": 1.9985988140106201,
-      "learning_rate": 2.2857142857142858e-05,
-      "loss": 9.9777,
+      "epoch": 0.8247422680412371,
+      "grad_norm": 2.2182443141937256,
+      "learning_rate": 2.1666666666666667e-05,
+      "loss": 10.0716,
       "step": 5
     },
     {
-      "epoch": 0.768,
-      "grad_norm": 2.318725109100342,
-      "learning_rate": 2.1428571428571428e-05,
-      "loss": 10.0346,
+      "epoch": 0.9896907216494846,
+      "grad_norm": 2.14205265045166,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 9.8673,
       "step": 6
     },
     {
-      "epoch": 0.896,
-      "grad_norm": 2.3330538272857666,
-      "learning_rate": 1.9999999999999998e-05,
-      "loss": 10.0432,
+      "epoch": 1.1546391752577319,
+      "grad_norm": 2.237016439437866,
+      "learning_rate": 1.8333333333333336e-05,
+      "loss": 9.9386,
       "step": 7
     },
     {
-      "epoch": 1.024,
-      "grad_norm": 2.386380672454834,
-      "learning_rate": 1.8571428571428572e-05,
-      "loss": 9.9037,
+      "epoch": 1.3195876288659794,
+      "grad_norm": 2.582629919052124,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 9.9123,
       "step": 8
     },
     {
-      "epoch": 1.152,
-      "grad_norm": 2.525782346725464,
-      "learning_rate": 1.7142857142857142e-05,
-      "loss": 9.8192,
+      "epoch": 1.4845360824742269,
+      "grad_norm": 2.809176445007324,
+      "learning_rate": 1.5e-05,
+      "loss": 9.9636,
       "step": 9
     },
     {
-      "epoch": 1.28,
-      "grad_norm": 2.6987812519073486,
-      "learning_rate": 1.5714285714285715e-05,
-      "loss": 9.7915,
+      "epoch": 1.6494845360824741,
+      "grad_norm": 2.4164328575134277,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 9.7264,
       "step": 10
     },
     {
-      "epoch": 1.408,
-      "grad_norm": 2.681642532348633,
-      "learning_rate": 1.4285714285714285e-05,
-      "loss": 9.6911,
+      "epoch": 1.8144329896907216,
+      "grad_norm": 2.5362374782562256,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 9.7569,
       "step": 11
     },
     {
-      "epoch": 1.536,
-      "grad_norm": 2.9579291343688965,
-      "learning_rate": 1.2857142857142857e-05,
-      "loss": 9.6579,
+      "epoch": 1.9793814432989691,
+      "grad_norm": 2.85573148727417,
+      "learning_rate": 9.999999999999999e-06,
+      "loss": 9.6862,
       "step": 12
     },
     {
-      "epoch": 1.6640000000000001,
-      "grad_norm": 3.073641300201416,
-      "learning_rate": 1.1428571428571429e-05,
-      "loss": 9.6782,
+      "epoch": 2.1443298969072164,
+      "grad_norm": 2.901190757751465,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 9.7247,
       "step": 13
     },
     {
-      "epoch": 1.792,
-      "grad_norm": 3.179450035095215,
-      "learning_rate": 9.999999999999999e-06,
-      "loss": 9.5699,
+      "epoch": 2.3092783505154637,
+      "grad_norm": 3.0155856609344482,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 9.6249,
       "step": 14
     },
     {
-      "epoch": 1.92,
-      "grad_norm": 3.1641924381256104,
-      "learning_rate": 8.571428571428571e-06,
-      "loss": 9.5342,
+      "epoch": 2.4742268041237114,
+      "grad_norm": 3.1418240070343018,
+      "learning_rate": 4.9999999999999996e-06,
+      "loss": 9.6827,
       "step": 15
     },
     {
-      "epoch": 2.048,
-      "grad_norm": 3.2276086807250977,
-      "learning_rate": 7.142857142857143e-06,
-      "loss": 9.5571,
+      "epoch": 2.6391752577319587,
+      "grad_norm": 3.1131372451782227,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 9.5966,
       "step": 16
     },
     {
-      "epoch": 2.176,
-      "grad_norm": 3.0624403953552246,
-      "learning_rate": 5.7142857142857145e-06,
-      "loss": 9.4244,
+      "epoch": 2.804123711340206,
+      "grad_norm": 3.168818712234497,
+      "learning_rate": 1.6666666666666667e-06,
+      "loss": 9.5602,
       "step": 17
     },
     {
-      "epoch": 2.304,
-      "grad_norm": 3.3281595706939697,
-      "learning_rate": 4.2857142857142855e-06,
-      "loss": 9.4145,
-      "step": 18
-    },
-    {
-      "epoch": 2.432,
-      "grad_norm": 3.577859878540039,
-      "learning_rate": 2.8571428571428573e-06,
-      "loss": 9.499,
-      "step": 19
-    },
-    {
-      "epoch": 2.56,
-      "grad_norm": 3.7756576538085938,
-      "learning_rate": 1.4285714285714286e-06,
-      "loss": 9.4644,
-      "step": 20
-    },
-    {
-      "epoch": 2.6879999999999997,
-      "grad_norm": 3.3084795475006104,
+      "epoch": 2.9690721649484537,
+      "grad_norm": 3.1356351375579834,
       "learning_rate": 0.0,
-      "loss": 9.348,
-      "step": 21
+      "loss": 9.592,
+      "step": 18
     },
     {
-      "epoch": 2.6879999999999997,
-      "step": 21,
-      "total_flos": 247581948423984.0,
-      "train_loss": 9.755998475211006,
-      "train_runtime": 4065.4818,
-      "train_samples_per_second": 0.092,
-      "train_steps_per_second": 0.005
+      "epoch": 2.9690721649484537,
+      "step": 18,
+      "total_flos": 212383098180048.0,
+      "train_loss": 9.849881066216362,
+      "train_runtime": 1793.8931,
+      "train_samples_per_second": 0.162,
+      "train_steps_per_second": 0.01
     }
   ],
   "logging_steps": 1.0,
-  "max_steps": 21,
+  "max_steps": 18,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 500,
@@ -182,7 +161,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 247581948423984.0,
+  "total_flos": 212383098180048.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null